(29g) Multiscale Modeling and Control of Cell Wall Thickness in Batch Pulp Digester

In pulping processes, the cell wall thickness (CWT) of wood chips is gradually decreased due to the removal of lignin, which causes the cellulose fiber to collapse. Paper made of collapsed and uncollapsed fibers has very different paper properties (e.g., density, surface smoothness and ink holdout capability). For example, the collapsed fiber is suitable for high-quality printing paper as it has dense and smooth surface whereas the uncollapsed fiber is better for making cartonboard that is resistant to bending and tearing [1-2]. However, the evolution of this important microscopic paper property is not able to be described by existing mathematical models as they only focus on macroscopic properties such as the concentration profiles of components and the temperature profiles of the cooking liquor and wood chip [3-6].

Motivated by this limitation, we developed a novel multiscale model by combining the mass continuity and thermal energy balance equations adopted from a modified “extended Purdue model” [6] with a kinetic Monte Carlo algorithm [7-9] to describe the microscopic events such as the evolution of CWT and Kappa number (i.e., residual lignin content in the wood pulp). To handle the high computational cost of the proposed multiscale mode, a data-driven reduced-order model was developed by utilizing the multivariable output error state space (MOESP) algorithm and the high-fidelity input/output data [10]. In addition, as the primary measurements of the process (i.e., Kappa number and CWT) are not available in real-time, a soft sensing system (i.e., Kalman filter) was developed to estimate the primary measurements by utilizing the secondary measurements (e.g., active effective alkali and dissolved lignin concentrations in the free-liquor phase) [11]. By taking advantages of the soft sensing system, a model-based feedback control framework was developed to regulate both the CWT and Kappa number of wood chips. Specifically, a model predictive control framework is employed to find the optimal manipulated input sequence (i.e., the free-liquor temperature of pulp digester) that drives the controlled output variables (i.e., primary measurements) to the pre-specified target values. The closed-loop simulation results demonstrated the proposed control framework outperforms other existing control strategies in regulating the microscopic paper properties.

References

[1] Johansson A. Correlations between fibre properties and paper properties. Master's thesis, KTH Royal Institute of Technology, Stockholm, Sweden, 2011.

[2] Facada MJ. Influence of Kraft paper quality on the performance of an industrial paper impregnation process. Ph.D. thesis, Universidade Tecnica de Lisboa, Lisboa, Portugal, 2015.

[3] Funkquist J. Grey-box identification of a continuous digester – a distributed-parameter process. Control Eng. Pract., 1997, 5, 919-930.

[4] Wisnewski PA.; Doyle FJ. Fundamental continuous pulp-digester model for simulation and control. AIChE J., 1997, 43, 3175-3192.

[5] Smith C. Studies of the mathematical modeling simulation and control of the operation of Kamyr continuous digester for the Kraft process. Ph.D. thesis, Purdue University, West Lafayette, IN, 1974.

[6] Bhartiya S.; Dufour P.; Doyle FJ. Fundamental thermal-hydraulic pulp digester model with grade transition. AIChE J., 2003, 49, 411-425.

[7] Lee D.; Mohr A.; Kwon JSI. Kinetic Monte Carlo modeling of multivariant binding of CTB proteins with GM1 receptor. Comp. Chem. Eng., 2018, 118, 283-295.

[8] Kwon JSI.; Nayhouse M.; Christofides PD.; Orkoulas G. Protein crystal shape and size control in batch crystallization: Comparing model predictive control with conventional operating policies. Ind. Eng. Chem. Res., 2014, 53, 5002-5014.

[9] Choi HK.; Kwon JSI. Multiscale modeling and control of Kappa umber and porosity in a batch-type pulp digester. AIChE J. (In press), 2019.

[10] Chou CT.; Verhaegen M. Subspace algorithms for the identification of multivariable dynamic errors-in-variables model. Automatica, 1997, 33, 1857-1869.

[11] Galicia HJ.; He QP.; Wang JA. A reduced order soft sensor approach and its application to a continuous digester. J. Process Control., 2011, 21, 489-500.